592

REVIEW ARTICLE Physical Examination of the Knee: A Review of the Original Test Description and Scientific Validity of Common Orthopedic Tests Gerard A. Malanga, MD, Steven Andrus, MD, Scott F. Nadler, DO, James McLean, MD

ABSTRACT. Malanga GA, Andrus S, Nadler SF, McLean J. tion on the sensitivity and specificity along with other infor- Physical examination of the knee: a review of the original test mation about the validity of these tests in clinical practice. To description and scientific validity of common orthopedic tests. standardize how physical examinations are performed and Arch Phys Med Rehabil 2003;84:592-603. compared, they should follow the original description or agreed-on standards. In addition, the significance of a physical Objectives: To present the original descriptions of common examination finding must be understood to ensure that patients orthopedic physical examination maneuvers of the knee and with knee complaints are accurately diagnosed and properly then to review the literature to support the scientific validity of treated. these tests. Key Words: Anterior cruciate ligament; Knee; Menisci, Data Sources: MEDLINE (1970–2000) searches were per- tibial; Physical examination; Posterior cruciate ligament; Re- formed, as were reviews of various musculoskeletal examina- habilitation. tion textbooks that describe physical examination maneuvers of © 2003 by the American Congress of Rehabilitation Medi- the knee. These references were then reviewed for additional cine and the American Academy of Physical Medicine and references and crossed back to the original description (when Rehabilitation possible) of these named tests. Study Selection: All articles that discussed the sensitivity USCULOSKELETAL PROBLEMS are among the most and specificity of the physical examination maneuvers were 1 extracted. This information was reviewed for accuracy and Mcommon reasons for patient visits to physicians. Unfor- tunately, it seems that most physicians are not adequately then summarized. 2-5 Data Extraction: Multiple MEDLINE and text searches trained in the evaluation of many musculoskeletal problems. were performed by using the terms of the test maneuver, the The knee is particularly susceptible to traumatic injury because joint tested, and the term physical examination. Any article of its vulnerable location midway between the hip and the with this information was reviewed until the article describing ankle, where it is exposed to the considerable forces transmit- the original description was found. Articles dating from that ted through the distal extremity from the ground. A thorough original article to the present were reviewed for information on examination of all knee structures should be a part of every the sensitivity and specificity of the test. knee evaluation. To ensure the most accurate diagnosis possi- Data Synthesis: Literature reviewing the sensitivity and ble, it is crucial that testing maneuvers be performed correctly specificity of the tests reviewed is summarized in text and table and that the examiner is aware of each maneuver’s sensitivity, form. The Lachman test seems to be very sensitive and specific specificity, and limitations. We reviewed the literature to find for the detection of anterior cruciate ligament tears. For pos- the original descriptions of commonly used physical examina- terior cruciate ligament tears, the posterior drawer test is also tion maneuvers of the knee and then searched the literature very sensitive and specific and is enhanced with other tests, again for studies that examined the sensitivity and specificity of such as the posterior sag sign. For meniscal tears, the McMur- these tests. ray test is very specific but has a very low sensitivity, whereas joint line tenderness has fairly good sensitivity but lacks good METHODS specificity. Although collateral ligament testing seems to be MEDLINE (1970–2000) searches were performed, as were sensitive and specific, there is a lack of well-designed studies reviews of various musculoskeletal examination textbooks. that scientifically validate the sensitivity and specificity of Key words used included knee, examination, anterior cruciate these tests. Common tests for patellofemoral pain and patellar ligament, posterior cruciate ligament, medial collateral liga- instability lack sensitivity when correlated with pathologic ment, lateral collateral ligament, patellofemoral pain, patellar operative findings. dislocation, McMurray test, Lachman test, anterior drawer Conclusions: Most physical examination tests could be ref- test, posterior drawer test, pivot shift test, apply grind test, erenced back to an original description, with variable informa- bounce home test, and grind test. The articles were obtained from local medical libraries and, when necessary, through interlibrary loan. The articles were then reviewed for additional references and searched back to the original description (when From the Kessler Institute for Rehabilitation, West Orange, NJ (Malanga, Andrus); possible) of the named tests. All articles that discussed the and Department of Physical Medicine and Rehabilitation, Univ of Medicine and Dentistry of New Jersey–New Jersey Medical School, Newark, NJ (Malanga, Nadler, sensitivity and specificity of these tests were extracted. This McLean). information was reviewed for accuracy and then summarized. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any Tests for the Anterior Cruciate Ligament organization with which the authors are associated. Reprint requests to Gerard A. Malanga, MD, Spine, Sports and Occupational The anterior cruciate ligament (ACL) is among the main Rehabilitation, Kessler Institute for Rehabilitation, 1199 Pleasant Valley Way, West Orange, NJ 07052, e-mail: [email protected]. stabilizers of the knee, and injury to it often results in signif- 0003-9993/03/8404-7322$30.00/0 icant disability. Three of the most commonly applied tests to doi:10.1053/apmr.2003.50026 determine an ACL injury are the anterior drawer test, the

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 593

Table 1: ACL Tests

Test Description Reliability and Validity Tests Comments

Anterior The subject is supine, hip Harilainen11 350 acute knees evaluated, with 79 arthroscopically confirmed drawer flexed to 45° and knee Sensitivity: 41% acute ACL injures. Prospective study. test flexed to 90°. Examiner sits Sensitivity (under anesthesia): 86% on the subject’s foot, with Katz and Fingeroth12 Testing performed only with patient under anesthesia. hands behind the proximal Sensitivity: 22.2% (acute injuries) tibia and thumbs on the Sensitivity: 53.8% (chronic injuries) Retrospective study. tibial plateau. Anterior force Specificity: Ͼ97% (acute and chronic) Limited sample size: 9 acute ACL injuries and 12 chronic. applied to the proximal Jonsson et al14 107 patients, all with documented acute or chronic ACL tibia. Hamstring tendons Sensitivity: 33% (acute injuries) ruptures. palpated with index fingers Sensitivity: 95% (chronic injuries) to ensure relaxation. Specificity not assessed because only positive ACL ruptures Increased tibial included. displacement compared Donaldson et al15 Retrospective study. with the opposite side is Sensitivity: 70% (acute injuries) Study not designed to evaluate specificity because authors indicative of an ACL tear. Sensitivity (under anesthesia): 91% reviewed only positive cases. Specificity: not reported Mitsou and Vallianatos16 144 knees, with 60 acute injuries all assessed within 3 days of Sensitivity: 40% (acute injuries) injury. In the group of 80 chronic injuries, the 4 false- Sensitivity: 95.24% (chronic injuries) negative drawer tests were associated with bucket-handle Specificity: not reported tears. Kim and Kim13 Testing performed only with patient under anesthesia. Sensitivity (under anesthesia): 79.6% Specificity. Not reported Retrospective study. Lachman Patient lies supine. Torg et al17 All ACL injuries were chronic. 93 knees with combined tears of test Knee held between full Sensitivity: 95% the ACL and median meniscus. All 5 false negatives were extension an 15° of flexion. Specificity: not reported associated with bucket-handle tears of the meniscus.

Donaldson et al15 Retrospective study. Sensitivity: 99% Study not designed to evaluate specificity because authors Femur is stabilized with 1 Specificity: Not reported reviewed only positive cases. hand while firm pressure is Katz and Fingeroth12 Testing performed only with patient under anesthesia. applied to the posterior Sensitivity (under anesthesia): aspect of the proximal tibia 84.67% in an attempt to translate it Specificity (under anesthesia): 95% Retrospective study. anteriorly. Limited sample size: 9 acute ACL injuries and 12 chronic injuries. Test is positive (indicating Kim and Kim13 Testing performed only with patient under anesthesia. ACL rupture) when there is Sensitivity (under anesthesia): 98.6% anterior translation of the Specificity: not reported Retrospective review study. tibia with “soft” endpoint. Mitsou and Vallianatos16 All ACL injuries were chronic. 144 knees, with 60 acute injuries Sensitivity: 80% (acute injuries) all assessed within 3 days of injury. Sensitivity: 98.8% (chronic injuries) Jonsson et al14 107 patients, all with documented acute or chronic ACL Sensitivity: 87% (acute injuries) ruptures. Sensitivity: 94% (chronic injuries) Specificity not assessed because only positive ACL ruptures included. Pivot shift Leg picked up at the ankle; the Lucie et al35 50 knees tested. There was not an adequate sample size of test knee is flexed by placing Sensitivity: 95% intact ACLs to determine specificity. the heel of the hand behind Katz and Fingeroth12 Testing performed only with patients under anesthesia. the fibula. As the knee is Sensitivity: 98.4% Retrospective study. extended, the tibia is Specificity: Ͼ98% Limited sample size: 9 acute ACL injuries and 12 chronic. supported on the lateral Donaldson et al15 Retrospective study. side with a slight valgus Sensitivity: 35% Study not designed to evaluate specificity because authors strain. A strong valgus force Sensitivity (under anesthesia): 98% reviewed only positive cases. is placed on the knee by the upper hand. At approximately 30° of flexion, the displaced tibia will suddenly reduce, indicating a positive pivot shift test.

Arch Phys Med Rehabil Vol 84, April 2003 594 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga

Lachman test, and the pivot shift test. These tests are summa- of the examiner. With the patient’s knee held between full rized in table 1. extension and 15 degrees of flexion, the femur is stabilized Anterior drawer test. Although the anterior drawer test has with one hand while firm pressure is applied to the pos- been widely used in the diagnosis of ACL ruptures, its origin terior aspect of the proximal tibia in an attempt to translate remains obscure. According to Paessler and Michel,6 Paul it anteriorly. A positive test indicating disruption of the Segund described, as early as 1879, the “abnormal anterior- anterior of cruciate ligament is one in which there is posterior mobility” of the knee that is associated with ACL proprioceptive and/or visual anterior translation of the ruptures. George Noulis, who Paessler and Michel6 credited tibia in relation to the femur with a characteristic “mushy” with the earliest description of what we now call the Lachman or “soft” end point. This is in contrast to a definite “hard” test, also explained the drawer tests in large degrees of flexion end point elicited when the anterior cruciate ligament is (knee flexion Ն90°). In a translation of Noulis’s 1875 thesis, intact.17 published in the textbook Diagnostic Evaluation of the Knee, Numerous studies have examined the sensitivity and speci- by Strobel et al.7 Noulis describes the following test: ficity of the Lachman test, and other studies have compared its With the patient’s leg flexed, the thigh can be grasped with accuracy with that of the original anterior drawer test. Torg17 one hand at the lower leg with the other hand keeping the originally reported that in 88 (95%) of 93 individuals with thumbs to the front and fingers to the back. If the lower leg combined lesions involving the ACL and medial meniscus, the is held in this grip and then moved backwards and for- Lachman test was positive. The false-negative tests were at- wards, it will be seen that the tibia can be moved directly tributed to incarcerated bucket-handle tears that blocked for- backwards and forwards.7 ward translation of the tibia. Donaldson et al15 noted a sensi- Noulis observed a great deal of tibia displacement when both tivity of greater than 99% for this test and found it to be cruciate ligaments were severed. The assumption that a posi- relatively unaffected by associated ligamentous or meniscal tive anterior drawer test indicates a tear of the ACL was not injuries. This was in contrast to the significant variability found commonly accepted until much later.8,9 Increased anterior tibial with the anterior drawer test when it was used with patients displacement, when compared with the uninvolved side, is now with injuries to the secondary restraints of the knee. The supported as indicative of an ACL tear.10 sensitivity of the Lachman test has been reported to range from Some limitations of this test remain, with sensitivities re- 80% to 99%, with a specificity of 95%.12,13,15-18 The test has ported of 22.2% to 41% when it is performed in the alert therefore been found to be the most sensitive and specific test patient and ranges of 79.6% to 91% when performed with the for the diagnosis of ACL tears, especially in cases of acute patient under anesthesia.11-16 Differences in the tests’ accuracy injury. are also noted in people with acute versus chronic injury, with There are certain limitations to the test. Draper and Schul- a positive anterior drawer test present in 50% to 95% of chronic thies19 noted that the Lachman test is not easily performed by injuries.9,10 Anterior drawer test results may also be affected by examiners who have small hands or on patients with a large a concomitant injury. Konin10 found that 54% of those with no thigh girth.20,21 Various modifications of the Lachman test have other injuries, 67% with associated medial meniscal injuries, been proposed, including “prone,”“drop leg,” and “stabilized” 82% with associated lateral meniscal injuries, and 89% with Lachman tests.20,22-24 Few studies have been performed com- associated medial collateral ligament (MCL) injury tested pos- paring the sensitivity and specificity of these modified tests itive for ACL injury. Those data suggest that the anterior with those of the original Lachman test. drawer test becomes increasingly more sensitive as the second- Pivot shift test. The pivot shift is both a clinical phenom- ary restraints of anterior stability are lost. enon that results in a complaint of a giving way of the knee and False-negative anterior drawer tests in instances of isolated a physical sign than can be elicited on examination of the ACL tears may occur secondary to protective spasms of the injured knee.25 Hey Groves26 in 1920 and Palmer27 in 1938 hamstring muscles and the anatomic configuration of the fem- published photographs showing patients voluntarily producing oral condyle, whereas false-positive results may occur with what is now called the pivot shift phenomenon. The phenom- posterior cruciate ligament (PCL) insufficiency.16 With a PCL enon was characterized as an anterior subluxation of the lateral tear, sagging of the tibia may result in a false sense of its tibial plateau in relation to the femoral condyle when the knee neutral position, which results in a false sense of anterior approaches extension with reduction produced with knee flex- movement, when, in fact, the tibia is moving into its normal ion.25,28-31 The pivot shift is enhanced by the convexity of the neutral position. tibial plateau in the sagittal plane.32,33 The pivot shift test was Overall, there is wide variation in the reported sensitivities initially described25,34 as follows: of the anterior drawer test, with those performed with the The leg is picked up at the ankle with one of the exam- patient under anesthesia having a somewhat limited use in the iner’s hands, and if the patient is holding the leg in clinical setting. The test’s relatively low sensitivity for detect- extension, the knee is flexed by placing the heel of the ing ACL tears in the acute setting should serve as a caution to other hand behind the fibula over the lateral head of the examiners not to rule out an acute ACL injury solely on the gastrocnemius. As the knee is extended, the tibia is sup- basis of a negative anterior drawer test. Conversely, the spec- ported on the lateral side with a slight valgus strain applied ificity of the test is quite high, and, therefore, a positive anterior to it. In fact, this subluxation can be slightly increased by drawer would strongly suggest ACL pathology. subtly internally rotating the tibia, with the hand that is Lachman test. The Lachman test was described by Torg et cradling the foot and ankle. A strong valgus force is placed al,17 who trained under Lachman at Temple University. Inter- on the knee by the upper hand. This impinges the subluxed estingly, Paessler and Michel6 traced descriptions of what we tibial plateau against the lateral femoral condyle, jamming now call the Lachman test to Noulis’s 1875 thesis. Despite the two joint surfaces together, preventing easy reduction these very early descriptions, the test was not widely recog- as the tibia is flexed on the femur. At approximately 30 nized or used until Torg’s classic description of the Lachman degrees of flexion, and occasionally more, the displaced test, which is given below: tibial plateau will suddenly reduce in a dramatic fashion. The examination is performed with the patient lying su- At this point, the patient will jump and exclaim, ‘that’s pine on the table with the involved extremity on the side it!’25

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 595

Table 2: PCL Tests

Reliability and Validity Test Description Tests Comments

Posterior sag sign Patient lies supine with the hip flexed Rubinstein et al40 Double-blinded, randomized, controlled to 45° and the knee flexed to 90°.In Sensitivity: 79% study. this position, the tibia rocks back, or Specificity: 100% 39 subjects enrolled in study (75 knees sags back, on the femur if the PCL is for analysis). torn. Normally, the medial tibial Examiners all fellowship-trained in plateau extends 1cm anteriorly sports medicine, with at least 5y beyond the femoral condyle when experience. the knee is flexed 90°. If this step-off Included only patients with chronic PCL is lost, the step-off test is tears. considered positive. Posterior drawer Subject is supine with the test hip Rubinstein et al40 See above. test flexed to 45°, knee flexed to 90°, and Sensitivity: 90% foot in a neutral position. The Specificity: 99% examiner is sitting on the subject’s foot with both hands behind the Loos et al39 Compilation study from registry of subject’s proximal tibia and thumbs Sensitivity: 51% knee surgeries in the United States on the tibial plateau. Apply a Specificity: not reported and Australia. 102 PCL injuries posterior force to the proximal tibia. included. Multiple examiners at Increased posterior tibial different sites, without indication that displacement as compared with the study was randomized or controlled. uninvolved side is indicative of a Moore and Larson50 Retrospective study of 20 patients. partial or complete tear of the PCL. Sensitivity: 67% All false negatives were found to have Specificity: not reported both anterior and posterior cruciate injuries at surgery. Hughston et al53 54 acute PCL tears studied over a 10-y Sensitivity: 55.5% period. Posterior drawer test was Specificity: not reported performed with patient under anesthesia. Clendenin et al49 Retrospective study of only 10 patients. Sensitivity: 100% Specificity: not reported Harilainen11 Prospective study that included only 9 Sensitivity: 90% patients with arthroscopically Specificity: not reported confirmed PCL tears. Quadriceps active Subject is supine with knee flexed to Daniel et al43 92 subjects included in study, 25 with test 90° in the drawer-test position. The Sensitivity: 98% no history of knee injury. foot is stabilized by the examiner, Specificity: 100% Study was not blinded or randomized, and the subject is asked to slide the as the examiners were told which foot gently down the table. knee was the index knee. Contraction of the quadriceps Rubinstein et al40 Double-blinded, randomized, controlled muscle in the PCL-deficient knee Sensitivity: 54% study. results in an anterior shift of the Specificity: 97% 39 subjects enrolled in study (75 knees tibia of Ն2mm. The test is for analysis). qualitative. Examiners all fellowship-trained in sports medicine, with at least 5y experience. Included only patients with chronic PCL tears.

Several studies have been performed to determine the diag- of an ACL tear. Additionally, a positive pivot shift test in a nostic sensitivity and specificity of the pivot shift test in the conscious patient may reflect the patient’s inability to protect diagnosis of ACL injuries. The reported sensitivity of pivot the knee, which may suggest that these patients are less likely shift in ACL injuries ranges from 84% to 98.4%, with a to respond to nonoperative treatment. specificity of 98.4% when the test is performed with the patient under anesthesia; in the alert patient, values as low as 35% have been described.12,15,35 Tests for the PCL Many authors8,11,36-38 have recommended various modifica- tions of the classic pivot shift test, including the addition of hip Many authors39-43 have reported on the difficulties associated abduction, knee flexion, and external tibial rotation. Because with the diagnosis of PCL injuries. Three tests commonly used the specificity is high, the pivot shift will usually be indicative to diagnose PCL injuries are the posterior sag sign, the poste-

Arch Phys Med Rehabil Vol 84, April 2003 596 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga rior drawer test, and the quadriceps active test. These tests are patients with chronic PCL tears; therefore, the accuracy of the summarized in table 2. posterior drawer test in acute PCL injuries cannot be inferred Posterior sag sign. Mayo Robson44 in 1903 described this from this study. Loos et al39 identified 102 PCL injuries in phenomenon, although it is unclear who coined the term pos- 13,316 knee operations performed in the United States and terior sag sign.45 A detailed description of the test as it is Australia. In their study, the sensitivity of the posterior drawer performed today is as follows: test was 51%. The study was not designed to evaluate speci- The patient lies supine with the hip flexed to 45 degrees ficity, because only patients with surgically documented PCL and the knee flexed to 90 degrees. In this position, the tibia injuries were included (no control group was reported). Hugh- “rocks back,” or sags back, on the femur if the posterior ston et al51 reported a sensitivity of 55.5% when the posterior cruciate ligament is torn. Normally, the medial tibial pla- drawer test was performed with the patient under anesthesia. teau extends 1 cm anteriorly beyond the femoral condyle The large number of false negatives was explained by a lack of when the knee is flexed 90 degrees. If this step off is lost, 46 injury to the posteriorly situated arcuate complex. Studies by it is considered positive for a posterior cruciate tear. Clendenin et al49 and Harilainen11 of patients under anesthesia Few studies have attempted to establish the posterior sag found positive posterior drawer tests even though posterior sign’s sensitivity and specificity in the diagnosis of PCL inju- 40 capsules were intact. ries. Rubinstein et al, in a blinded, randomized, and con- Simonsen et al41 found a sensitivity of 91% and a specificity trolled study to assess the accuracy of the clinical examination of 80%, and O’Shea et al42 found a sensitivity of 100% and a in PCL injuries, reported a sensitivity of the posterior sag sign specificity of 99% when performing a complete knee exami- for detecting PCL injury of 79%, with a specificity of 100%. nation; neither reported on the specificity and sensitivity of the The overall sensitivity of the clinical examination to detect individual maneuvers for PCL tears. In summary, the posterior PCL injury when all tests were used was 90%, with a speci- 47 drawer test has a high sensitivity and specificity, and its accu- ficity of 99%. Staubli and Jakob, in a nonrandomized, un- racy is increased when results are combined with other tests for blended, and controlled study, evaluated the accuracy of the posterior instability, such as the posterior sag sign. “gravity sign near extension,” in which the knee is maintained Quadriceps active test. The quadriceps active test was in near extension. They reported that the gravity sign near described by Daniel et al43 in 1988. Daniel’s description of the extension was detectable in 20 of 24 PCL-deficient knees, for test follows: a sensitivity of 83%. Overall, the posterior sag sign is a useful With the subject supine, the relaxed limb is supported with test for diagnosing PCL injuries, with a relatively high sensi- the knee flexed to 90 degrees in the drawer-test position. tivity and very high specificity when used on its own. The subject should execute a gentle quadriceps contraction Posterior drawer test. Noulis accurately described the op- to shift the tibia without extending the knee. At this posing forces of the ACL and PCL in his 1875 thesis.6,7 6 90-degree angle, the patellar ligament in the normal knee Paessler and Michel, in their historical review, give the fol- is oriented slightly posterior and contraction of the quad- lowing account of Noulis’s description: riceps does not result in an anterior shift of the tibia When the leg was then moved forward and backward, it although there may be a slight posterior shift. If the was found that the tibia will slide anteriorly and posteri- posterior cruciate ligament is ruptured, the tibia sags into orly. Noulis observed much movement of the tibia when posterior subluxation and the patellar ligament is then both cruciates had been severed. When only the ACL was directed anteriorly. Contraction of the quadriceps muscle severed, movement of the tibia could be shown when the in the posterior cruciate-ligament deficient knee results in knee was “barely flexed.” However, when the posterior an anterior shift of the tibia of 2 mm or more. The test is cruciate ligament had been divided, it took about 110 43 6 qualitative. degrees of flexion to produce this movement of the tibia. 43 6 In an unblinded, nonrandomized study, Daniel reported a Paessler and Michel explained that Noulis’s 110° of flexion positive quadriceps active test in 41 of 42 knees that had a PCL would translate into 70° of flexion today, because at that time rupture, for a sensitivity of 98%. Daniel reported a negative 180° was considered as full extension. A more detailed and quadriceps active test in all the normal knees and the knees contemporary description of the posterior drawer test as it is with ACL disruptions but intact PCLs, for a specificity of commonly performed follows: 100%. Rubinstein40 reported a sensitivity for the quadriceps The subject is supine with the test hip flexed to 45 degrees, active test of 54% and a specificity of 97%. This compares with knee flexed to 90 degrees, and foot in neutral position. The findings of a sensitivity of 79% and 90% for the posterior sag examiner is sitting on the subject’s foot with both hands sign and posterior drawer test, respectively, with Ն99% spec- behind the subject’s proximal tibia and thumbs on the ificity for both tests. tibial plateau. Apply a posterior force to the proximal tibia. Significantly different sensitivities for the quadriceps active Increased posterior tibial displacement, as compared to the test may reflect study methodology and slightly different pa- uninvolved side, is indicative of a partial or complete tear 48 tient populations. The blinded, randomized, and controlled of the PCL. 40 8,11,39,40,43-50 study by Rubinstein did not find the quadriceps active test to Numerous studies have reported on the accuracy be as sensitive in detecting PCL disruption as were other tests. of the posterior drawer test in identifying injuries to the PCL. Many of these studies lacked adequate sample sizes or had other flaws in their methodology, which makes interpretation Tests for the Medial and Lateral Collateral Ligaments of the results difficult. Rubinstein40 reported that the posterior The MCL is among the most frequently injured ligaments in drawer test was the most accurate test for identifying PCL the knee. Valgus stress testing is the primary method used to injuries, with a sensitivity of 90% and a specificity of 99%. In diagnose MCL injury, although few studies have evaluated its general, when all of the clinical tests for PCL injuries were accuracy or interexaminer reliability. Injuries of the lateral analyzed on the basis of the grade of PCL tear, the examination collateral ligament (LCL) are less common, and even fewer sensitivity for grade I sprains was only 70%, with a 99% studies have evaluated the accuracy of the varus stress test in specificity, whereas the sensitivity of grade II and III sprains the diagnosis of this injury. These tests are summarized in was 97%, with 100% specificity. The study included only table 3.

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 597

Table 3: MCL and LCL Tests

Reliability and Validity Text Description Tests Comments

Valgus stress Patient supine on the examination table. Harilainen11 72 patients studied with MCL tears confirmed test Flex the knee to 30° over the side of the Sensitivity: 86% on arthroscopy. Valgus stress testing was table, place 1 hand about the lateral Specificity: not reported performed in 20° of flexion, and testing in aspect of the knee, and grasp the ankle extension was not done. Clinical examination with the other hand. Apply abduction was performed in the ER under unknown (valgus) stress to the knee. The test conditions, and no indication is given must also be performed in full regarding the number of examiners or their extension. training. There is also no documentation of the elapsed time between ER evaluation and arthroscopic evaluation. Garvin et al56 Retrospective study of 23 patients who had Sensitivity: 96% undergone surgery for MCL tears. Specificity: not reported Nonstandardized clinical examination of the MCL was used, sometimes with the patient under anesthesia and sometimes performed before anesthesia. McClure et al57 Physicians did not perform testing in this study, Interexaminer reliability and the physical therapists’ experience was in extension: 68% varied. Standardized examination techniques Interexaminer reliability between the examiners were not used. Data in 30°flexion: 56% variability of the clinical categories was Sensitivity: not reported insufficient to allow for accurate Specificity: not reported determination of reliability values. Varus stress Patient supine on the examination table. Harilainen11 Only 4 patients were studied with LCL tears test Flex the knee to 30° over the side of the Sensitivity: 25% confirmed on arthroscopy. Varus stress table, place 1 hand about the medial Specificity: not reported testing was performed in 20° of flexion, and aspect of the knee, and grasp the ankle testing in extension was not done. Clinical with the other hand. Apply adduction examination was performed in the ER under (varus) stress to the knee. The test must unknown conditions, and no indication was also be performed in full extension. given regarding the number of examiners or their training. There was also no documentation of the elapsed time between ER evaluation and arthroscopic evaluation.

Abbreviation: ER, emergency room.

Valgus and varus stress tests. Although the originator of valgus stress test positive at 0° indicates a tear of both the PCL the valgus and varus stress tests for detecting ligament laxity is and the medial-compartment ligaments. They did not find that unclear, Palmer9 described “abduction and adduction rocking” the integrity of the ACL had any effect on the valgus stress test of the knee to determine the integrity of the collateral liga- in extension. Marshall and Rubin52 noted that the valgus stress ments, which is an early reference to the valgus and varus test in extension implicates one or both cruciate ligaments in stress tests used today. A description of Palmer’s test from his addition to the MCL and posterior capsule. 1938 article is given below: Injuries resulting in straight lateral instability are rare. Hugh- In order to demonstrate lateral rocking, it is of the greatest ston et al53 reported on the operative findings of 3 patients with importance that the patient be made to relax his muscles. straight lateral instability demonstrated by positive varus stress In many cases this can be performed if the extremity is testing in extension. Surgery revealed a torn PCL, lateral cap- grasped so, that it rests firmly and painlessly in the grip of sule, arcuate ligaments, and a torn ACL in 2 patients, among the examiner. The best way is to hold the leg with the foot 52 supported in the armpit with the calf resting against the other findings. Marshall and Rubin, in their review of this forearm. The other hand supports the back of the knee. subject, reported that a positive varus stress test in flexion When it is felt that the muscles are relaxed, a surprise implicates the LCL, whereas a positive test in extension de- abduction movement is made. It is then felt how the notes a combined injury of the LCL, popliteus, and cruciate articular surfaces snap apart and, when the muscles start to ligaments. function as a reflex action, spring together again with a Little is known about the accuracy of valgus and varus stress click which is clearly discernible to the hand supporting testing. Harilainen11 noted that of 72 patients with arthroscopi- the back of the knee. cally confirmed MCL tears, 62 were diagnosed on clinical Currently, varus and valgus testing are performed with slight examination, for a sensitivity of 86%. Of 4 patients with an abduction of the hip and 30° of knee flexion. Hughston et al8 LCL tear confirmed on arthroscopy, 1 instability was diag- concluded that a valgus stress test positive at 30° and negative nosed on clinical examination, for a sensitivity of 25%. The at 0° indicates a tear limited to the medial-compartment liga- flawed design of the study significantly limits its overall clin- ments (MCL with or without the posterior capsule), whereas a ical utility.

Arch Phys Med Rehabil Vol 84, April 2003 598 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga

Table 4: Patellofemoral Tests

Test Description Reliability and Validity Tests Comments Patellofemoral The subject is lying supine with the No studies were found that The diagnosis of PFS is based on clinical grinding test knees extended. The examiner stands document the sensitivity or examination, including the patella and next to the involved side and places specificity of the compression test. The lack of another patellofemoral the web space of the thumb on the patellofemoral grinding test criterion standard (eg, arthroscopy) in the compression superior border of the patella. The in the diagnosis of PFS. diagnosis of PFS makes any test (for PFS) subject is asked to contract the determination of sensitivity or specificity quadriceps muscle while the of specific clinical tests for this condition examiner applies downward and problematic. inferior pressure on the patella. Pain with movement of the patella or an inability to complete the test is indicative of patellofemoral dysfunction. Apprehension Performed by pressing on the medial Sallay et al72 19 patients underwent arthroscopic test (for aspect of the patella with the knee Sensitivity: 39% evaluation in this study, and all 19 patella flexed 30° with the quadriceps Specificity: not reported exhibited gross lateral laxity of the dislocation) relaxed. It requires the thumbs of patellofemoral articulation under both hands pressing on the medial anesthesia. This laxity was most side of the patella to exert the prominent at 70° to 80° of flexion. laterally directed pressure. Often the finding is surprising to the patient, and he/she becomes uncomfortable and apprehensive as the patella reaches the point of maximum passive displacement, with the result that he/she begins to resist and attempts to straighten the knee, thus pulling the affected patella back into a relatively normal position.

Abbreviation: PFS, patellofemoral syndrome.

A few studies have compared the clinical examination of Tests for Patellofemoral Disorders collateral ligament injuries with varus and valgus stress testing Anterior knee pain is among the most common complaints versus magnetic resonance imaging (MRI) of the liga- that cause patients to consult a physician. Two of the most 54,55 54 ments. Yao et al reported that the agreement between common tests used in the evaluation of patellofemoral disor- MRI and the clinical grade of injury was modest, with a 65% 55 ders are the patellar compression, or grinding, test and the correct classification. Mirowitz and Shu reported a correla- patella apprehension test. These tests are summarized in table tion coefficient between MR diagnosis and clinical diagnosis of 4. .73 for MCL injuries. Garvin et al56 reported in a retrospective Patellofemoral grinding test. The term chondromalacia study that a tear of the MCL was predicted from the clinical 59 examination in 22 of 23 patients, for a sensitivity of 96%. The patellae did not appear in published form until 1924. Aleman 60 retrospective design of the study, which included only serious is credited with using this term as early as 1917, although the injuries, along with the nonstandardized clinical examination first mention of chondromalacia in the English language was in performed primarily with patients under anesthesia, limits its 1933 by Kulowski.61 In 1936, Owre62 published the results of clinical usefulness. a clinical and pathologic investigation of the patella. His de- McClure et al57 addressed the interexaminer reliability of the scription of the patellofemoral grinding test follows: valgus stress test. Tests were performed by 3 physical thera- Pressure-pain over the patella is tested by clasping the pists on 50 patients with unilateral knee problems. The inter- patella with the thumb and index finger of each hand with examiner reliability was 0.6, with 68% agreement between the remaining fingers resting against the thigh and leg. examiners, for the knee in extension and .16, with 56% agree- While the patient lies with the leg relaxed and extended, ment between examiners, for the knee in 30° of flexion. the patella is pressed against the medial and lateral femoral In summary, there is a lack of well-designed studies that condyles. By moving the patella in an upward and down- evaluate the sensitivity and specificity of the varus and valgus ward direction the greater part of the surface cartilage may stress tests, or their interexaminer reliability, in the diagnosis be examined in this manner. In some cases, pain is elicited and grading of collateral ligament injuries.58 Future studies on the slightest pressure of the patella against the condyle, would be clinically useful but challenging to design, because at other times considerable pressure must be exerted to most patients with collateral ligament injuries are currently obtain a positive response of an unpleasant sensation.62 managed without surgery. Therefore, the criterion standard of Owre considered a positive test as indicated by pain to be arthroscopically identifying ligament injuries would be hard to predictive of pathologic changes to the retropatellar cartilage or justify. chondromalacia patella. In current use, a positive test may or

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 599 may not be associated with the pathologic diagnosis of chon- examiner can almost dislocate the patella over the lateral dromalacia patella, as determined by direct arthroscopic visu- femoral condyle. Often the finding is surprising to the alization and probing. Solomon et al5 provide a more contem- patient and he becomes uncomfortable and apprehensive porary description of the test: “The subject is lying supine with as the patella reaches the point of maximum passive dis- the knees extended. The examiner stands next to the involved placement, with the result that he begins to resist and side and places the web space of the thumb on the superior attempts to straighten the knee, thus pulling the affected border of the patella. The subject is asked to contract the patella back into a relatively normal position.71 quadriceps muscle, while the examiner applies downward and Sallay et al,72 in their 1996 study, reported on the charac- inferior pressure on the patella.” Pain with movement of the teristic clinical and arthroscopically determined pathologic patella, or an inability to complete the test, is indicative of findings associated with patellar dislocations. Only 39% of patellofemoral dysfunction. patients with a history of dislocation were found to have a These 2 descriptions vary significantly. In the first (or pas- positive apprehension sign. In contrast, 83% had a moderate to sive grind) test, the potentially pathologic and painful patella is large effusion, and 70% had significant tenderness over the compressed against the patellofemoral joint in an effort to posterior medial soft tissues. MRI revealed a moderate to large reproduce pain from pathologic chondral changes on the pa- effusion on all scans. Increased signal adjacent to the adductor tella. The latter description involves an active contraction of tubercle was seen in 96% of patients, tearing of the medial the quadriceps with external compression on the patella, which patellofemoral ligament was found in 87%, and increased sig- presumably would result in a more dynamic compression of the nal was noted in the vastus medialis oblique muscle in 78%. On patella against the femur. Neither test replicates how the patella arthroscopic evaluation, gross lateral laxity of the patellofemo- normally moves within the patellofemoral joint. ral articulation of all subjects was most prominent at 70° to 80° There are no studies that document the sensitivity or the of flexion. This degree of flexion is significantly higher than the specificity of the patellofemoral grinding test in the diagnosis 30° classically recommended for the apprehension sign and of patellofemoral syndrome. Several studies in the last 20 may explain the low sensitivity of this test in the diagnosis of years, however, have shown a generally poor correlation be- patella dislocation.72 tween retropatellar pain and articular cartilage damage.63-69 We caution against the use of the active grind test because we have Tests for Meniscal Injuries found it to cause pain in normal asymptomatic subjects while Meniscal tears occur commonly; however, their clinical di- teaching this maneuver to students and residents. agnosis is often difficult, even for an experienced clinician. O’Shea et al42 reported on the diagnostic accuracy of clinical Because the menisci are avascular and have no nerve supply on examination of the knee in patients with arthroscopically doc- their inner two thirds, an injury to the meniscus can result in umented knee pathology, including chondromalacia patella. little or no pain or swelling, which makes accurate diagnosis They reported that only 11 of 29 patients were correctly diag- even more challenging. In 1803, Hey73 described “internal nosed as having the pathologic findings of chondromalacia derangement of the knee,” and since then a significant literature patella on the basis of history, physical examination, and stan- on the clinical diagnosis of meniscal tears has evolved. These dard radiographs, for a sensitivity of 37%. tests are summarized in table 5. In summary, the correlation is poor between the clinical Joint line tenderness. Joint line palpation is among the history and examination and the diagnosis of chondromalacia most basic maneuvers, yet it often provides more useful infor- patella. An explanation for the disparity between clinical signs mation than the provocative maneuvers designed to detect and pathologic findings is not simple and reflects the continu- meniscal tears. Flexion of the knee enhances palpation of the ing question of what ultimately causes the pain in patients with anterior half of each meniscus. The medial edge of the medial patellofemoral syndrome. meniscus becomes more prominent with internal rotation of the Apprehension test for patellar dislocation. This test was tibia, allowing for easier palpation. Alternatively, external ro- first described by Fairbank in 1936.70 The test, often referred to tation allows improved palpation of the lateral meniscus. as the Fairbank’s Apprehension Test, has been described by The literature notes a sensitivity for joint line tenderness of Fairbank as such: 55% to 85%, with a specificity range of 29% to 67%.74,75 Thus, While examining cases of suspected recurrent dislocation joint line tenderness is likely to be present in those with of the patella, I have been struck by the marked apprehen- meniscal tears. However, joint line tenderness alone is common sion often displayed by the patient when the patella is to other diagnoses and is not pathognomonic for meniscal pushed outwards in testing the stability of this bone. Not injury. uncommonly, the patient will seize the examiner’s hands McMurray test. The McMurray test is among the primary to check the manipulation, which she finds uncomfortable clinical tests to evaluate for a meniscal tear. McMurray76 first and regards as distinctly dangerous. This sign, when described the test in 1940.77 The original description of the test, present, I regard as strong evidence in favour of a diag- as described by McMurray, was: nosis of slipping patella.70 In carrying out the manipulation with patient lying flat, the A more detailed and more recent description of the appre- knee is first fully flexed until the heel approaches the hension test for subluxation of the patella was given by Hugh- buttock; the foot is then held by grasping the heel and ston.71 His description is as follows: using the forearm as a lever. The knee being now steadied This test is carried out by pressing on the medial side of by the surgeon’s other hand, the leg is rotated on the thigh the patella with the knee flexed about 30 degrees and with with the knee still in full flexion. During this movement the quadriceps relaxed. It requires the thumbs of both the posterior section of the cartilage is rotated with the hands pressing on the medial side of the patella to exert the head of the tibia, and if the whole cartilage, or any frag- laterally directed pressure. Accordingly, the leg with mus- ment of the posterior section, is loose, this movement cles relaxed is allowed to project over the side of the produces an appreciable snap in the joint. By external examining table and is supported with the knee at 30 rotation of the leg the internal cartilage is tested, and by degrees of flexion by resting the leg on the thigh of the internal rotation any abnormality of the posterior part of examiner who is sitting on a stool. In this position the the external cartilage can be appreciated. By altering the

Arch Phys Med Rehabil Vol 84, April 2003 600 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga

Table 5: Tests for Meniscal Injuries

Reliability and Test Description Validity Tests Comments

Joint line The medial edge of the medial Kurosaka et al74 Prospective blinded study of 160 patients tenderness meniscus becomes more prominent Sensitivity: 55% with meniscal tears that were with internal rotation of the tibia, Specificity: 67% arthroscopically identified. Acute injuries allowing for easier palpation. were excluded. Alternatively, external rotation allows Fowler and Lubliner75 Prospective study of 160 patients (161 improved palpation of the lateral Sensitivity: 85% knees) with mensical tears that were meniscus. Specificity: 29.4% arthroscopically identified. Anderson and Lipscomb82 Prospective evaluation of 100 patients Sensitivity: 77% evaluated by 1 examiner. Specificity: Not reported McMurray With the patient lying flat, the knee is Evans et al77 Prospective study of 104 patients. test first fully flexed; the foot is held by Sensitivity: 16% Interexaminer reliability between the 2 grasping the heel. The leg is rotated Specificity: 98% examiners of the study was only fair. on the thigh with the knee still in full Fowler and Lubliner75 Prospective study of 160 patients (161 flexion. By altering the position of Sensitivity: 29% knees) with mensical tears that were flexion, the whole of the posterior Specificity: 95% arthroscopically identified. segment of the cartilages can be Kurosaka et al74 Prospective blinded study of 160 patients examined from the middle to their Sensitivity: 37% with meniscal tears that were posterior attachment. Bring the leg Specificity: 77% arthroscopically identified. Acute injuries from its position of acute flexion to a were excluded. right angle while the foot is retained Anderson and Lipscomb82 Prospective evaluation of 100 patients first in full internal rotation and then Sensitivity: 58% evaluated by 1 examiner. in full external rotation. When the Specificity: Not reported click occurs (in association with a torn meniscus), the patient is able to state that the sensation is the same as he/she experienced when the knee gave way previously. Apley grind The patient is prone. The surgeon Fowler and Lubliner75 Prospective study of 160 patients (161 test grasps 1 foot in each hand, externally Sensitivity: 16% knees) with mensical tears who were rotates as far as possible, then flexes Specificity: 80% arthroscopically identified. both knees together to their limit. The Kurosaka et al74 Prospective blinded study of 160 patients feet are then rotated inward and Sensitivity: 13% with meniscal tears who were knees extended. The surgeon then Specificity: 90% arthroscopically identified. Acute injuries applies his left knee to the back of the were excluded. patient’s thigh. The foot is grasped in both hands, the knee is bent to a right angle, and powerful external rotation is applied. Next, the patient’s leg is strongly pulled up, with the femur being prevented from lifting off the couch. In this position of distraction, external rotation is repeated. The surgeon (next) leans over the patient and compresses the tibia downward. Again he rotates powerfully, and, if addition of compression produces an increase in pain, this grinding test is positive and meniscal damage is diagnosed. Bounce The test is performed with the patient No studies were found home supine and the foot cupped in the that identified the test examiner’s hand. With the patient’s accuracy of this specific knee completely flexed, the knee is test. passively allowed to extend. The knee should extend completely or bounce home into extension with a sharp endpoint. If extension is not complete or has a rubbery end feel, there is probably a torn meniscus or some other blockage.

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 601

position of flexion of the joint the whole of the posterior Bounce home test. The bounce home test is designed to segment of the cartilages can be examined from the middle evaluate a lack of full extension in the knee, which may to their posterior attachment.... Probably the simplest indicate a torn meniscus or other pathology, such as a loose routine is to bring the leg from its position of acute flexion body or a joint effusion. The test is performed with the patient to a right angle, whilst the foot is retained first in full supine, with his/her foot cupped in the examiner’s hand. With internal, and then in full external rotation.... When the the patient’s knee completely flexed, the knee is passively click occurs with a normal but lax cartilage, the patient allowed to extend. The knee should extend completely or experiences no pain or discomfort, but when produced by bounce home into extension with a sharp endpoint. If extension a broken cartilage, which has already given trouble, the is not complete or has a rubbery end feel, there is probably a patient is able to state that the sensation is the same as he torn meniscus or some other blockage present.44 experienced when the knee gave way previously.76 Oni83 described a modification of the bounce home test, Several studies75,78-82 have been performed to determine the clinical accuracy of the McMurray test in predicting meniscal which he labeled the knee-jerk test, in which the knee is pathology. Four studies74,75,77,82 were found that evaluated the forcibly extended in 1 quick jerk and pain occurs in the region 84 McMurray test as it was originally described, without modifi- of tissue injury. Shybut and McGinty described the forced cation. hyperextension test of the knee, which, in contrast to the There seems to be a wide variation in the reported sensitiv- bounce home test and the jerk test, involves forced hyperex- ities (16%–58%) and specificities (77%–98%) of the McMur- tension of an already extended knee. A block to full extension ray test for detecting meniscal tears.74,75,77,82 Evans et al77 indicates a positive test and may indicate a meniscal tear. found low agreement between examiners, and the McMurray Fowler and Lubliner,75 in their study on the predictive value of test was not found useful for diagnosing lateral meniscal tears. 5 clinical tests for meniscal pathology, reported a sensitivity of Overall, these findings support the continued utility of the 44% and a specificity of 95% for the forced hyperextension of McMurray test in combination with other physical examination the knee test. tests and in patients with a history that is suggestive of menis- cal involvement. The test should not be overly emphasized DISCUSSION when it is negative, given its low diagnostic sensitivity. Apley grind test. The Apley grind test was described by Physical examination is essential in the diagnosis of any Apley in 1947.79,80 The original description of the test follows: medical problem, including musculoskeletal pathology. Given For this examination the patient lies on his face. He should the large number of patients who consult physicians because of be on a couch not more than 2 feet high, or the tests musculoskeletal complaints, it is important that physicians become difficult, and he must be well over to the edge of truly understand how to perform the various musculoskeletal the couch nearest the surgeon. To start the examination, tests that are commonly used in clinical practice and that they the surgeon grasps one foot in each hand, externally ro- understand the significance of the test results. For physicians to tates as far as possible, and then flexes both knees together communicate unambiguously, both clinically and in the scien- to their limit. When this limit has been reached, he tific literature, the physical examination must be standardized. changes his grasp, rotates the feet inward, and extends the The information presented here is somewhat artificial, in that knees together again....Thesurgeon then applies his left physical examination tests are generally not performed in iso- knee to the back of the patient’s thigh. It is important to lation, but together with other tests. It seems that this would observe that in this position his weight fixes 1 of the levers increase the diagnostic accuracy of the physical examination, absolutely. The foot is grasped in both hands, the knee is which is important in developing a treatment plan and mini- bent to a right angle, and the powerful external rotation is mizing unnecessary diagnostic testing. In addition, clinicians applied. This test determines whether simple rotation pro- often find that several structures are injured. The pathologic duces pain. Next, without changing the position of the hands, the patient’s leg is strongly pulled upward, while changes in 1 structure may alter how a different structure is the surgeon’s weight prevents the femur from rising off examined. For example, a meniscal tear in conjunction with an the couch. In this position of distraction, the powerful ACL tear may alter the findings on anterior drawer and Lach- external rotation is repeated. Two things can be deter- man tests. Also, inflammatory changes and swelling after in- mined: (1) whether or not the maneuver produces pain and jury may result in a decreased ability to detect findings on (2), still more important, whether the pain is greater than physical examination. Nonetheless, we believe that a properly in rotation alone without the distraction. If the pain is performed physical examination is invaluable in the optimal greater, the distraction test is positive, and a rotation sprain evaluation and treatment of patients with musculoskeletal in- may be diagnosed. juries and should be taught and performed in a standardized Then the surgeon leans well over the patient and, with fashion, with a clear understanding of the significance of each his whole body weight, compresses the tibia downward test maneuver. onto the couch. Again he rotates powerfully, and if addi- The Lachman test seems to be very sensitive and specific for tion of compression had produced an increase of pain, this the detection of ACL tears. For PCL tears, the posterior drawer grinding test is positive, and meniscal damage is diag- test is also very sensitive and specific and is enhanced with nosed.80 other tests, such as the posterior sag sign. For meniscal tears, Fowler and Lubliner,75 in their report of 5 clinical signs for the McMurray test is very specific but has a very low sensi- meniscal pathology, prospectively evaluated the accuracy of tivity, whereas joint line tenderness has fairly good sensitivity the Apley grind test. They reported an overall sensitivity of but lacks good specificity. Although collateral ligament testing 16% and a specificity of 80%. Kurosaka et al74 noted a sensi- seems to be sensitive and specific, there is a lack of well- tivity of 13% and a specificity of 90%, with an overall accuracy designed studies that scientifically validate the sensitivity and of 28%, for the Apley grind. The results of these prospective specificity of these tests. Common tests for patellofemoral pain studies show the limited predictive value of the Apley grind and patellar instability lack sensitivity when correlated with test for the diagnosis of meniscal injuries. pathologic operative findings.

Arch Phys Med Rehabil Vol 84, April 2003 602 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga

CONCLUSION 25. Galway HR, MacIntosh DL. The lateral pivot shift: a symptom The importance of a properly performed physical examina- and sign of anterior cruciate ligament insufficiency. Clin Orthop tion of the knee cannot be overemphasized. In the same light, 1980;Mar-Apr(147):45-50. 26. Hey Groves EW. The crucial ligaments of the knee joint: their the diagnostic accuracy and limitations of the various tests for function, rupture and the operative treatment of the same. Br J knee pathology need to be understood. With this knowledge, Surg 1920;7:505-15. the thoughtful clinician is better able to plan both diagnostic 27. Johnson RJ. The anterior cruciate: a dilemma in sports medicine. and treatment strategies for knee injuries. Int J Sports Med 1982;3:71-9. 28. Galway HR, Beaupre A, MacIntosh DL. Pivot shift: a clinical sign References of symptomatic anterior cruciate insufficiency. J Bone Joint Surg 1. Boskey AL. Musculoskeletal disorders and orthopedic conditions. Br 1972;54:763-4. JAMA 2001;285:619-23. 29. Kennedy JC. Anterior subluxation of the lateral tibial plateau. In: 2. Ahern MJ, Soden M, Schultz D, Clark M. The musculo-skeletal James SL, editor. Late reconstructions of injured ligaments of the examination: a neglected clinical skill. AustNZJMed1991;21: knee. New York: Springer-Verlag; 1978. p 94-8. 303-6. 30. Losee RE, Johnson TR, Southwick WO. Anterior subluxation of 3. Clawson DK, Jackson DW, Ostergaard DJ. It’s past time to reform the musculoskeletal curriculum. Acad Med 2001;76:709-10. the lateral tibial plateau. A diagnostic test and operative repair. 4. Freedman KB, Bernstein J. The adequacy of medical school J Bone Joint Surg Am 1978;60:1015-30. education in musculoskeletal medicine. J Bone Joint Surg Am 31. Slocum DB, Larson RL, James JL. Late reconstruction procedures 1998;80:1421-7. used to stabilize the knee. Orthop Clin North Am 1973;4:679-89. 5. Solomon DH, Simel DL, Bates DW, Katz JN, Schaffer JL. Does 32. Kujala UM, Nelimarkka O, Koskinen SK. Relationship between this patient have a torn meniscus or ligament of the knee? Value the pivot shift and the configuration of the lateral tibial plateau. of the physical examination. JAMA 2001;286:1610-20. Arch Orthop Trauma Surg 1992;111:228-9. 6. Paessler HH, Michel D. How new is the Lachman test? Am J 33. Larson RL. Physical examination in the diagnosis of rotatory Sports Med 1992;20:95-8. instability. Clin Orthop 1983;Jan-Feb(172):38-44. 7. Strobel M, Stedtfeld HW, Feagin JA, Telger TC. Diagnostic 34. MacIntosh DL, Galway HR. The lateral pivot shift. A symptom- evaluation of the knee. New York: Springer-Verlag; 1990. atic and clinic sign of anterior cruciate insufficiency. Paper pre- 8. Hughston JC, Andrews JR, Cross MJ, Moschi A. Classification of sented at: Annual Meeting of the American Orthopaedic Associ- knee ligament instabilities. Part I: the medial compartment and ation: 1972 June 28; Tucker’s Town (Bermuda). cruciate ligaments. J Bone Joint Surg Am 1976;58:159-72. 35. Lucie RS, Wiedel JD, Messner DG. The acute pivot shift: clinical 9. Palmer I. On injuries to the ligaments of the knee joint. A clinical correlation. Am J Sports Med 1984;12:189-91. study. Acta Chir Scand Suppl 1938;53:282. 36. Bach BR Jr, Warren RF, Wickiewicz TL. The pivot shift phenom- 10. Konin JG. Special tests for orthopedic examination. Thorofare enon: results and description of a modified clinical test for anterior (NJ): SLACK; 1997. cruciate ligament insufficiency. Am J Sports Med 1988;16:571-6. 11. Harilainen A. Evaluation of knee instability in acute ligamentous injuries. Ann Chir Gynaecol 1987;76:269-73. 37. Jakob RP, Staubli HU, Deland JT. Grading the pivot shift. Ob- 12. Katz JW, Fingeroth RJ. The diagnostic accuracy of ruptures of the jective tests with implications for treatment. J Bone Joint Surg Br anterior cruciate ligament comparing the Lachman test, the ante- 1987;69:294-9. rior drawer sign, and the pivot shift test in acute and chronic knee 38. Noyes FR, Grood ES, Cummings JF, Wroble RR. An analysis of injuries. Am J Sports Med 1986;14:88-91. the pivot shift phenomenon. The knee motions and subluxations 13. Kim SJ, Kim HK. Reliability of the anterior drawer test, the pivot induced by different examiners. Am J Sports Med 1991;19:148- shift test, and the Lachman test. Clin Orthop 1995;Aug(317):237- 55. 42. 39. Loos WC, Fox JM, Blazina ME, Del Pizzo W, Friedman MJ. 14. Jonsson T, Althoff B, Peterson L, Renstrom P. Clinical diagnosis Acute posterior cruciate ligament injuries. Am J Sports Med of ruptures of the anterior drawer ligament: a comparative study of 1981;9:86-92. the Lachman test and the anterior drawer sign. Am J Sports Med 40. Rubinstein RA Jr, Shelbourne KD, McCarroll JR, VanMeter CD, 1982;10:100-2. Rettig AC. The accuracy of the clinical examination in the setting 15. Donaldson WF III, Warren RF, Wickiewicz T. A comparison of of posterior cruciate ligament injuries. Am J Sports Med 1994; acute anterior cruciate ligament examinations. Initial versus ex- 22:550-7. amination under anesthesia. Am J Sports Med 1985;13:5-10. 41. Simonsen O, Jensen J, Mouritsen P, Lauritzen J. The accuracy of 16. Mitsou A, Vallianatos P. Clinical diagnosis of ruptures of the clinical examination of injury to the knee joint. Injury 1984;16: anterior cruciate ligament: a comparison between the Lachman 96-101. test and the anterior drawer sign. Injury 1988;19:427-8. 42. O’Shea KJ, Murphy KP, Heekin RD, Herzwurm PJ. The diagnos- 17. Torg JS, Conrad W, Kalen V. Clinical diagnosis of anterior tic accuracy of history, physical examination, and radiographs in cruciate ligament instability in the athlete. Am J Sports Med the evaluation of traumatic knee disorders. Am J Sports Med 1976;4:84-93. 1996;24:164-7. 18. DeHaven KE. Arthroscopy in the diagnosis and management of 43. Daniel DM, Stone ML, Barnett P, Sachs R. Use of the quadriceps the anterior cruciate ligament deficient knee. Clin Orthop 1983; active test to diagnose posterior cruciate-ligament disruption and Jan-Feb(172):52-6. measure posterior laxity of the knee. J Bone Joint Surg Am 19. Draper DO, Schulthies SS. Examiner proficiency in performing 1988;70:386-91. the anterior drawer and Lachman tests. J Orthop Sports Phys Ther 44. Mayo Robson AW. Ruptured crucial ligaments and their repair by 1995;22:263-6. operation. Ann Surg 1903;37:716-8. 20. Draper DO. A comparison of stress tests used to evaluate the 45. Barton TM, Torg JS, Das M. Posterior cruciate ligament insuffi- anterior cruciate ligament. Phys Sportsmed 1990;18:89-96. ciency. A review of the literature. Sports Med 1984;1:419-30. 21. Draper DO, Schulthies SS. A test for eliminating false positive 46. Magee DJ. Orthopedic physical assessment. 3rd ed. Philadelphia: anterior cruciate ligament injury diagnoses. J Athl Train 1993;28: WB Saunders; 1997. p 506-98. 355-7. 47. Staubli HU, Jakob RP. Posterior instability of the knee near 22. Feagin JA, Cooke TD. Prone examination for anterior cruciate extension. A clinical and stress radiographic analysis of acute ligament insufficiency. J Bone Joint Surg Br 1989;71:863. injuries of the posterior cruciate ligament. J Bone Joint Surg Br 23. Wroble RR, Lindenfeld TN. The stabilized Lachman test. Clin 1990;72:225-30. Orthop 1988;Dec(237):209-12. 48. Hughston JC. The absent posterior drawer test in some acute 24. Adler GG, Hoekman RA, Beach DM. Drop leg Lachman test. A posterior cruciate ligament tears of the knee. Am J Sports Med new test of anterior knee laxity. Am J Sports Med 1995;23:320-3. 1988;16:39-43.

Arch Phys Med Rehabil Vol 84, April 2003 PHYSICAL EXAMINATION TESTS OF THE KNEE, Malanga 603

49. Clendenin MB, DeLee JC, Heckman JD. Interstitial tears of the 65. Dehaven KE, Dolan WA, Mayer PJ. Chondromalacia patellae in posterior cruciate ligament of the knee. Orthopedics 1980;3:764- athletes. Clinical presentation and conservative management. 72. Am J Sports Med 1979;7:5-11. 50. Moore HA, Larson RL. Posterior cruciate ligament injuries. Re- 66. Hvid I, Anderson LI. The quadriceps angle and its relation to sults of early surgical repair. Am J Sports Med 1980;8:68-78. femoral torsion. Acta Orthop Scand 1982;53:577-9. 51. Hughston JC, Bowden JA, Andrews JR, Norwood LA. Acute tears 67. Kelly MA, Insall JN. Historical perspectives of chondromalacia of the posterior cruciate ligament. Results of operative treatment. patellae. Orthop Clin North Am 1992;23:517-21. J Bone Joint Surg Am 1980;62:438-50. 68. Leslie IJ, Bentley G. Arthroscopy in the diagnosis of chondroma- 52. Marshall JL, Rubin RM. Knee ligament injuries—a diagnostic and lacia patellae. Ann Rheum Dis 1978;37:540-7. 69. Tria AJ Jr, Palumbo RC, Alicea JA. Conservative care for patel- therapeutic approach. Orthop Clin North Am 1977;8:641-68. lofemoral pain. Orthop Clin North Am 1992;23:545-54. 53. Hughston JC, Andrews JR, Cross MJ, Moschi A. Classification of 70. Fairbank HA. Internal derangement of the knee in children and knee ligament instabilities. Part II. The lateral compartment. adolescents. Proc R Soc Med 1936;30:427-32. J Bone Joint Surg Am 1976;58:173-9. 71. Hughston JC. Subluxation of the patella. J Bone Joint Surg Am 54. Yao L, Dungan D, Seeger LL. MR imaging of tibial collateral 1968;50:1003-26. ligament injury: comparison with clinical examination. Skeletal 72. Sallay PI, Poggi J, Speer KP, Garrett WE. Acute dislocation of the Radiol 1994;23:521-4. patella. A correlative pathoanatomic study. Am J Sports Med 55. Mirowitz SA, Shu HH. MR imaging evaluation of knee collateral 1996;24:52-60. ligaments and related injuries: comparison of T1-weighted, T2- 73. Hey W. Practical observations in surgery. Philadelphia: James weighted, and fat-saturated T2-weighted sequences—correlation Humphreys; 1805. with clinical findings. J Magn Reson Imaging 1994;4:725-32. 74. Kurosaka M, Yagi M, Yoshiya S, Muratsu H, Mizuno K. Efficacy 56. Garvin GJ, Munk PL, Vellet AD. Tears of the medial collateral of the axially loaded pivot shift test for the diagnosis of a meniscal ligament: magnetic resonance imaging findings and associated tear. Int Orthop 1999;23:271-4. injuries. Can Assoc Radiol J 1993;44:199-204. 75. Fowler PJ, Lubliner JA. The predictive value of five clinical signs in 57. McClure PW, Rothstein JM, Riddle DL. Intertester reliability of the evaluation of meniscal pathology. Arthroscopy 1989;5:184-6. clinical judgments of medial knee ligament integrity. Phys Ther 76. McMurray TP. The semilunar cartilages. Br J Surg 1942;29:407-14. 1989;69:268-75. 77. Evans PJ, Bell GD, Frank C. Prospective evaluation of the Mc- 58. Snyder-Mackler L. Valgus stress test study [letter]. Phys Ther Murray test. Am J Sports Med 1993;21:604-8. 78. Stratford PW, Binkley J. A review of the McMurray test: defini- 1990;70:204. tion, interpretation, and clinical usefulness. J Orthop Sports Phys 59. Dugdale TW, Barnett PR. Historical background: patellofemoral Ther 1995;22:116-20. pain in young people. Orthop Clin North Am 1986;17:211-9. 79. Gillis L. Diagnosis in orthopaedics. Toronto: Butterworth; 1969. 60. Karlson S. Chondromalacia patellae. Acta Chir Scand 1939;83: 80. Gould JA, Dabies GJ. Orthopaedic and sports physical therapy. 347-81. Toronto: CV Mosby; 1985. 61. Kulowski J. Chondromalacia of the patella. Fissural cartilage 81. Hoppenfeld S. Physical examination of the spine and extremities. degeneration; traumatic chondropathy: report of three cases. Norwalk (CT): Appleton-Century-Crofts; 1976. JAMA 1933;100:1837-40. 82. Anderson AF, Lipscomb AB. Clinical diagnosis of meniscal tears. 62. Owre A. Chondromalacia patellae. Acta Chir Scand 1936;77 Description of a new manipulative test. Am J Sports Med 1986; Suppl 41:1-159. 14:291-3. 63. Abernethy P, Wilson G, Logan P. Strength and power assessment. 83. Oni AO. The knee jerk test for diagnosis of torn meniscus [letter]. Issues, controversies and challenges. Sports Med 1995;19:401-17. Clin Orthop 1985;Mar(193):309. 64. Darracott J, Vernon-Roberts B. The bony changes in chondroma- 84. Shybut GT, McGinty JB. The office evaluation of the knee. lacia patellae. Rheumatol Phys Med 1971;11:175-9. Orthop Clin North Am 1982;13:497-509.

Arch Phys Med Rehabil Vol 84, April 2003